Use of real-time polymerase chain reaction (rtPCR) as a diagnostic tool for influenza infection in a vaccine efficacy trial.

BACKGROUND: Conventional techniques for diagnosing influenza based on viral cell culture or disease serology have limitations, and molecular assays, such as real-time polymerase chain reaction (rtPCR) are increasingly used. OBJECTIVES: To evaluate the use of rtPCR as a diagnostic tool for the determination of influenza virus infection. STUDY DESIGN: This prospective, double-blind, placebo-controlled, randomised efficacy study was conducted in persons aged 18-64 years. Cases of influenza-like-illness (ILI), defined as at least one systemic symptom [fever ≥37.8°C and/or myalgia] and at least one respiratory symptom [cough and/or sore throat] were identified by active and passive surveillance. For each case of suspected ILI, nasal and throat swabs were collected and analysed by viral culture and rtPCR. RESULTS: 227 ILI cases were positive by rtPCR while 64% (145/227) were positive by both rtPCR and culture. For both assays, the maximum percentage of swabs that tested positive was on Day 0, thereafter positive samples by rtPCR remained constant until Day 5 but decreased progressively by culture. All rtPCR positive cases with a viral load of below 4.5log(10) copies/sample were negative by culture. There were however culture negative cases with high viral loads. Vaccine efficacy for influenza was estimated as 54.7% by rtPCR (culture positive or negative) and 61.6% by culture irrespective of match to vaccine strain. Clinical severity was not significantly different between culture positive cases and culture negative but rtPCR positive cases. CONCLUSIONS: rtPCR is a sensitive and specific diagnostic tool for influenza vaccine efficacy studies.